Since the launch of the generative artificial intelligence (AI) tool ChatGPT in November 2022, AI has taken the corporate training industry — and many others — by storm. With applications in content development, personalization, translation and more, it’s unlikely that the generative AI hype will die down any time soon. In fact, the corporate training market continues to see new AI-powered solutions designed to automate various learning and development (L&D) processes and tasks, freeing up valuable time and resources for busy training professionals.

Recent Training Industry research found that although many companies are using AI in some capacity, there’s a lack of understanding on how to implement it effectively. In other words, says Tom Whelan, Ph.D., director of corporate research at Training Industry, “We’re all jumping in the pool, but we’re not quite sure of our swimming abilities.” This may stem from the fact that most companies are focused on training employees to use AI to complete job-specific tasks in an effort to increase overall productivity, rather than training C-level leaders on strategic AI planning to ensure smooth long-term adoption.

L&D professionals can support more strategic AI adoption by asking thoughtful questions to determine if it makes business sense to use AI for an L&D-specific task or process. Here, we’ll outline four questions you should ask to determine when it makes sense to use AI for training.

1.     Which L&D processes are candidates for automation?

AI can automate many different L&D tasks and processes. Training Industry research found that the three most common AI use cases for L&D are: personalization, simulations and localization. For example, AI could tailor an existing leadership training program for first-time leaders by including relevant examples and emphasizing foundational leadership skills for new managers. AI can also personalize training for leaders in different industries.

Graham Glass, founder and CEO of CYPHER Learning, explains that today’s personalized learning — if you’re using generative AI — is “built specifically just for you.” Thus, AI can deliver “cutting-edge” personalization in a matter of minutes.

Another way AI can personalize learning is by creating curated learning pathways, says Apratim Purakayastha (AP), general manager, enterprise solutions and chief product and technology officer at Skillsoft. Imagine if you’re a training professional tasked with curating a learning path for a first-time technical manager, and you have over 50,000 pieces of content at your disposal, he says. “AI can be assistive in curating learning experiences and learning paths.”

Content development is another popular use case for AI in L&D. Although Glass says you wouldn’t want to use AI to create 100% of a course, you can use it to create 80% of a course in just a few minutes. Then, you can fine-tune it as needed.

It’s worth noting that just because personalization, simulations and localization are the three most common use cases for AI in L&D, your organization’s use cases will vary depending on the organization’s maturity in using AI and its specific needs, Whelan says.

Identifying L&D processes (i.e., administration, content development, translation, etc.) that could be automated is the first step in leveraging AI for greater L&D efficiency.

2.     Which AI tools do you have access to?

Next, it’s important to consider which AI tools you have access to. Does your organization have any policies in place that may prevent you from using free AI tools like ChatGPT, Gemini or Copilot? If not, using a free tool is a good place to start. Paid versions often come with advanced features that free versions may not offer, such as enhanced customization options, data privacy controls and the ability to integrate seamlessly within a learning management system (LMS).

If you work in a regulated industry or if your stakeholders have data privacy concerns, it may make sense to invest in a paid version if you have the budget to do so. Before purchasing, attend a demo or take advantage of free trial periods to explore the platform’s capabilities.

It’s also critical to ask vendors security-related questions to ensure your data and your learners’ information will be protected, says Purakayastha. Questions like, “How secure is the learning environment? What data is being stored, and how is it used? What personal information could get leaked?” are important to determine whether a vendor’s solution meets your organization’s privacy standards.

If you don’t have the budget to invest in a paid tool but your stakeholders are still concerned about data privacy issues, you may want to start using AI for “small, low-stakes tasks,” such as outlining job aids, Whelan says. This will help your stakeholders become more comfortable with the technology over time. After all, “Your stakeholders are resistant now, but it doesn’t necessarily mean they’ll always be resistant.”

3.     Do employees have the AI and data skills needed to use AI effectively?

It’s likely that employees in your organization would benefit from foundational AI and/or data skills training. In fact, LinkedIn Learning’s 2024 Workplace Learning Report found that 4 in 5 people want to learn more about how to use AI in their profession.

Here are some ways to build employees’ AI and data skills:

• Teach them to write effective prompts: An effective prompt is specific, uses unambiguous language and offers the necessary context. Through trial and error, and lots of practice, employees can refine their prompt writing skills for more meaningful interactions with AI tools.
• Reinforce the importance of human oversight: The critical thinking and intent behind human responses is a major advantage over AI outputs, Whelan says. Whether you’re using AI to translate content or to create a new course from the ground up, it’s crucial to maintain human oversight over AI-generated content.
• Build data privacy skills: Ensure that employees understand the importance of keeping personal and organizational data secure when using AI tools. Employees should know how to handle data responsibly, identify potential privacy risks and follow relevant regulations.
• Teach them to recognize AI hallucinations/misinformation: AI tools can occasionally generate inaccurate or misleading information, known as “hallucinations.” Educate employees on how to recognize these inaccuracies by cross-checking AI outputs with reliable sources.

4. Does your organization have the infrastructure needed to support an AI tool?

Implementing AI into your L&D processes requires more than just purchasing the right tool. It also requires having the proper information technology (IT) infrastructure in place to support it, Whelan explains. First, your organization will need the necessary computing power to handle AI workloads, especially if the tool requires high processing speeds or the ability to process large datasets. Cloud-based AI solutions can be a good option for organizations that don’t have the in-house infrastructure for AI.

Then, assess whether your current technologies can integrate with the AI tool. Make sure your IT team has the bandwidth to support these integrations and that your existing platforms are compatible with the AI tool you’re considering.

As AI tools and their applications evolve, it’s important that your infrastructure scales accordingly. This means investing in cloud solutions or adaptable on-premise systems that can accommodate future upgrades and increased AI usage.

Lastly, consider the training and support your IT team might need to maintain and troubleshoot the AI tool. Having a dedicated team that understands how to manage AI-powered technology will help you maximize its potential while minimizing disruptions.

Conclusion

Ultimately, “The whole training and learning industry will be reinvented because of AI,” Purakayastha says.

By considering the questions outlined above, you can ensure that you’re using AI for L&D thoughtfully and strategically — no matter how far along your organization is on its AI adoption journey.